The effects of inversed ratio ventilation (IRV) on arterial oxygenation during mechanical ventilation in patients with acute respiratory failure

Resuscitation.

22 (1991)

Elsevier Scientific

93-101

Publishers

93

Ireland

Ltd.

The effects of inversed ratio ventilation (IRV) on arterial oxygenation during mechanical ventilation in patients with acute respiratory failure Atsuo

Sari, Shigeki Yamashita, Ryuichi Kawata, Mitsuru

Department

of Anesthesia.

(Received

August

24th,

Kurashiki

Central

1990; Revision

Takashi Toriumi, Kenn Nakashima, Kunihiro and Akitomo Yonei Hospital,

received

1-1-I

March

Miwa,

Kurashiki,

Okayama

18th. 1991; Accepted

April

7!0 (Japan)

17th. 1991)

We investigated the effects of inversed ratio ventilation by altering the inspiratory: expiratory and assessing the time course changes in the intrapulmonary shunting ( Q,/ Q,) in I4 patients respiratory failure. Stepwise prolongation of the I:E ratio from I :I.9 to 2:1 and then to 2.6 applied when PEEP failed to raise the Pao, above 80 mmHg while breathing oxygen. A

(I:E) ratio with acute or 4:1 was significant

decreasein Qs/ Qt was observed following prolongation of the I:E ratio from I:!.9 (,Q,/ Q, = 45 f 9%) to 2:1 ( Qs/ Q, = 29 f 9%) but not with further prolongation of the I:E ratio ( Q,/ Qt = 27 * 7%). Improvement of the pulmonary ventilation/perfusion imbalance became more marked with continued IRV and a significant increase in Paoz was observed at 6 h after initiating prolongation of the inspiratory time (P < 0.05). There were no significant changes in hemodynamics, Pace,, or peak inspiratory pressure during IRV. This ventilatory pattern may be indicated when PEEP fails to improve Pao,, but prolongation of the inspiratory time above an I:E ratio of 2:1 did not produce a greater improvement in Q,/ Q, and further increases in Pao, did not occur after more than IO h of IRV in our I4 patients.

Respiration - Mechanical ventilation Oxygen transport - Tissue oxygenation

IRV - I:E ratio - Mean airway - Intrapulmonary shunt

pressure

-

Oxygenation

-

INTRODUCTION

Since Reynolds [I,21 demonstrated that mechanical ventilation with a high inspiration: expiration ratio (1:E) could be used to successfully treat infants with the idiopathic respiratory distress syndrome, inversed ratio ventilation (IRV) has been applied experimentally and clinically to acute respiratory failure (ARF). However, the improvement in arterial oxygenation achieved by IRV is controversial, with some investigators reporting a beneficial effect [3--51 while others have found it to be of no value [6-91. This discrepancy might be due to the different durations of IRV used or to differences in the subjects of these studies (i.e., variations in the animal species or in the type of human disease treated). Address all correspondence and reprint requests to: A. Sari, M.D.,

Hospital,

l-l-l

Miwa,

0300-9572/91/$03.50 Printed

and Published

Kurashiki, 0

Okayam

1991 Elsevier

in Ireland

Dept of Anesthesia,

710, Japan.

Scientific

Publishers

Ireland

Ltd.

Kurashiki

Central

94

We examined the effects of IRV on the time course of changes in the QJQt in 14 patients with ARF due to diverse causes, in whom positive end-expiratory pressure (PEEP) had failed to increase the Pa@. MATERIALS

AND METHODS

Patients Fourteen patients suffering from ARF were studied within 6 h of initiating mechanical ventilation in the intensive care unit (ICU). Approval for the investigation was obtained from the Ethics Committee of the hospital and informed written consent was obtained from the patients’ relatives. None of the patients had any history of lung disease. The causes of the pulmonary injury were diverse and all patients had bilateral diffuse infiltrative changes on chest radiographs. The severity of their disease was evaluated ai the time of admission to ICU using the APACHE II classification system [lo]. All patients underwent mechanical ventilation with PEEP and received infusions of an inotropic agent and/or vasodilators; the infusion rate remained unchanged during the study. They were sedated with intravenous diazepam or buprenorphine and paralyzed with pancuronium bromide to facilitate controlled mechanical ventilation with PEEP or IRV. Patients received blood, plasma expanders, and balanced electrolyte solutions as required by their overall clinical condition: vigorous diuresis was not attempted unless left ventricular failure was documented. According to the results of bacterial cultures of blood or sputum, patients received intravenous therapy with from one to several antibiotics. Apparatus Mechanical ventilation was provided using a Servo ventilator 900 C (SiemensElma AB, Sweden). The end-tidal CO2 concentration was monitored with a CO2 Analyzer 930 (Siemens Elema AB, Sweden), and inspired oxygen concentration (F1oZ) was regulated by a calibrated OrAir Mixer 961(Siemens-Elema AB, Sweden) within the range of 20.9-100% oxygen. The F102, peak inspiratory pressure (PIP), and mean airway pressure (MAWP) were monitored from a digital display mounted on ventilator. Arterial and venous pressures were measured at end-expiration using a pressure transducer (1290, Hewlett Packard, MA). The transducers were positioned at the mid-axillary line with the patients lying in the supine position, and the atmospheric pressure was used as the zero reference point. Ventilator settings The initial settings were as follows: tidal volume, 8-12 ml/kg body weight; respiratory rate, l&20 breaths/min (BPM); PEEP, 4 cmHzO (up to 6 cmH,O); and F102, 1.0. Peak inspiratory pressure ranged between 30 and 40 cmH20. When the Paoz could not be maintained above 80 mmHg with a PEEP of 6 cmHaO, the I:E ratio was increased stepwise from 1: 1.9 to 2:l and then to 2.6:1 or 4:l (with a PEEP of 6 cmH,O),while measuring the blood gases. Respiratory frequency was set at 12 BPM and the min volume was initially adjusted to produce a Pacoz of 4&50 mmHg. Prolongation of the inspiratory time was achieved by 67% prolongation at each step with a different pause time (0, 5, or 10%).

95

Study sequence

Measurements were made after 1 and 3 h at a given ventilator setting. The study was divided into three periods. The first was the control period involving controlled mechanical ventilation with a PEEP of 6 cmH,O. An inspiratory time of 25% and an inspiratory hold of 10% gave an 1:E ratio of 1:1.9. In the second period, IRV was applied with an 1:E ratio of 2: 1 for at least 3 h, and in the third period it was applied at a rate of 2.6:l or 4:1 for 48 h or more. Observations were made of the effect of IRV on Paoz and also on the progressive changes in PaOz with time. Measurements

After 1 and 3 h at a given ventilator setting, direct measurements of the following parameters were made: heart rate (HR), arterial blood pressure, pulmonary artery pressure (PAP), central venous pressure (CVP), pulmonary capillary wedge pressure (PCWP), mean airway pressure (MAWP), and PIP. The PO,, Pco*, and pH in arterial and mixed venous blood were measured with a blood gas analyzer (Model 178 pH/Blood Gas Analyzer, Corning Medical and Scientific, Copenhagen). A Swan Ganz triple-lumen, balloon-tipped, flow-directed thermodilution catheter (Edward Laboratories, MA) was positioned in the pulmonary artery via the right internal jugular vein to measure the right atria1 pressure (RAP), PAP, PCWP, and cardiac output (CO). The catheter position was confirmed by a portable chest radiograph and by visualization of the appropriate wave form in the pulmonary artery after wedging. Cardiac output was determined by the thermodilution technique using 10 ml of 5% dextrose in water at 0°C which was injected in the expiratory period. Measurements were performed in triplicate and averaged. Calculations

Pulmonary vascular resistance (PVR) was calculated from the ratio of CO to the difference between the mean PAP (MPA) and PCWP. Oxygen content was calculated from the hemoglobin oxygen-carrying capacity and the amount of dissolved oxygen, estimated from the Pa02 and oxygen solubility. The Pao;/FiC$ ratio was calculated.as arterial oxygen tension/inspired 02 concentration. The intrapulmonary shunt (Q,/Q,) was calculated using the standard shunt equation [11]: Q,lQ, = (Cc’oZ - Cao2)/(Cc’oz - CSo,), where CC/O* is the pulmonary capillary oxygen content, and Cao, is the arterial oxygen content. To calculate &/Qr we assumed that pulmonary capillary oxygen tension was the same as alveolar oxygen tension. Oxygen delivery (DO*) was calculated as the product of Cao, and CO. Oxygen consumption (VOW)was calculated using the Fick principle as follows: VO, = CO x (Cao, - C’;jo,). All variables were determined using a calculator (HC-20, Epson, Shinshu Seiki, Tokyo) with a program for oxygenation and hemodynamics (American Edwards, CA). Statistical analysis

The significance of differences

was assessed using analysis of variance and

P values of less than 0.05 were considered significant. All results are presented as

the mean f SD.

96

Table I.

Patient

characteristics. 14 66 zt 16

No. of patients Age (years) Sex Female Male Cause of ARF

8 6 5 3 4 1 1 19 f 3*

Chest infection Sepsis Heart failure Aspiration Unknown Duration of mechanical ventilation Duration of IRV (days) Complications during IRV

(days)

2

Pneumothorax Outcome (APACHE 11 score) Survived (16 f 5) Died (24 f 3)* *Significant

difference

99 4

6 8

from the patients

who died (P < 0.01). Values are the mean

* S.D.

RESULTS

The patient characteristics are shown in Table I. On the 2nd and 3rd days following IRV with an 1:E ratio of 2: 1, two patients with interstitial pneumonia developed a pneumothorax. A chest tube was inserted immediately and provided effective relief. The other patients had no complications related to the IRV itself. The APACHE II score ranged from 13 to 27 (mean f SD. = 20 f 5), and 8 patients had a score above 20. There was a significant difference in this score between the

Table II.

Effects of IRV on arterial

Parameter/I:E

ratio

oxygenation 1

: 1.9

with acute respiratory 2:

1.0

b’2

Paoz

mmHg

Pace, PH Base excess A-aDO

mmHg

QJQ,

%

Pao2/Fto2 PIP MAWP

cmH20 cmHzO

*Significant zt SD.

in patients

mEq/L mmHg

difference

65 51 7.41 8 581 45 65 25 10

from the values obtained

1

failure. 2.&4

1.0 f

15

14 * 0.13 f 8 f 65 *9 ?? 15 zk 5 *4

??

115

f

52 7.41 7 452 29 115 26 13

+

42;

13 0.12 * 4 ?? 130* zk 9* f 42’ + 7 f 5* ??

136

:1

I.&-O.7 zt 46*

51 7.38 5 461 27 136 33 17

?? 15 f 0.04 * 5 f 143* f 7+ zrz 46; f 9 + 7*

at an 1:E ratio of 1:1.9 (P < 0.05). Values are the mean

97 Table III.

Effects

respiratory

failure.

Parameter/I:E

of IRV on pulmonary

ratio

1

Cl SVI

Ilminlm2 ml/min/m2

MAP HR CVP SVR

mmHg beats/min

11 30 9 23 195 35 12 8 164 507

dyn . s/cm’ mmHg vol”/ vol”/

PVR PTo2 Cao2 C%+

ml/min/m2 mllminlm2

yo2 Do2

and oxygen

: 1.9

3.7 32 78 116 8 1136

mmHg dyn . s/cm5 mmHg g-m/m2/beat g-m/m2/beat mmHg

PCWP LVSWI RVSWI MPAP

hemodynamics

2: 1.7 + 14 & 15 jz 29 ztz 4 f 553

I

4.2 32 84 113 8 1227

??

zt 6 f 16 f 8 f 8 f 121 f 6 + 3 +z 3 f 79 f 253

delivery

10 31 10 24 223 38 13 9 175 591

in patients

with acute

:1

2.w zt 1.7 f 13 zt 17 * 20 f 3 f 497 f 4 ztz 18 ?? 7 f 8 + 113 zt 5 f 4 f 3 f 81 f 273

3.0 30 67 107 11 995 11 32 9~6 24 225 39 12 7 146 378

f

0.7 11 f 16 f 18 z& 3 f 243 f 3 zt 12

??

f f ??

f f f f

8 84 7

1 2 25 164

CI = cardiac index, SVI = stroke volume index, MAP = mean arterial pressure, HR = heart rate, CVP = central venous pressure, SVR = systemic vascular resistance, PCWP = pulmonary capillary wedge pressure, LVSWI = left ventricular stroke work index, RVSWI = right ventricular stroke work index, MPAP = mean pulmonary artery pressure, PVR = pulmonary vascular resistance, Pvo2 = mixed venous oxygen tension? Cao, = arterial oxygen content, Cvo2 = mixed venous oxygen content, Vo, =oxygen consumption, DO, = oxygen delivery. Values are the mean f S.D.

. 300 -

t*

y =7.71 x +67.49

.*

r =0.37 250

-

“=3g

P CO.05 200

PaO~/F,On

-

150 -

100 -

50 -

I

5

10

15

20 cmH,O

MAWP Fig. 1. Correlation between Pao,/Fto, and MAWP (x). Only 39 data points refer to-blood gas values measured simultaneouly with the MAWP.

are shown because

they

98 300 y = 76.418 (x t 1Y”” r = 0.927

250

200

PaOJ F,Oz 150 P < 0.01

100 P < 0.05

50

T

Pmean i c

1

3

5

10

15

SD

"24"* hrs.

Time Fig. 2. Changes in Paoz/F,o, with time (hours). The correlation between Pao,/F,o, (y) and time (x) is described by an exponential curve:y = 76.418 (X + 1)“.2”34, (r = 0.927, P < 0.01). The data shown was obtained at multiple I:E ratio.

patients (16 f 5) and those who died (24 f 3) (P < 0.01). Table II shows the 1:E ratios and concomitant blood gas values, the respiratory parameters and the PIP and MAWP data, while Table III shows systemic and . pulmonary hemodynamics and tissue oxygenation. A significant reduction in QJQ, associated with an increase in both Paoa and the Paoz/F& ratio was observed as the I:E ratio increased. There was a good correlation between &/&JJ) and the Pao2/F,02 ratio (x)0, = -0.111x + 45.6, n = 48, r = 0.46, P < 0.01). These changes in oxygenation corresponded quite closely to the changes in MAWP and a direct linear correlation was observed between MAWP(x) and the Paoz/F1oz ratio O~)(_Y = 7.71x + 67.49, r = 0.37, P C 0.05, n = 39) (Fig. 1). The improvement noted became even greater with continuing IRV and a significant improvement in oxygenation was noted at just over 3 h from the initiation of IRV (Fig. 2). No further increase in Pa@ was observed after 10 h. There were no significant changes in pulmonary or systemic hemodynamics, tissue oxygenation, or oxygen delivery parameters. Peak inspiratory pressure did not increase significantly with the increase in the I:E ratio and no significant changes in Pace, were observed at any I:E ratio used. surviving

DISCUSSION

The rn?joT finding of this study was that prolongation of the inspiratory time improved QJQ, in 14 mechanically ventilated patients with ARF induced by diverse causes, who were poorly oxygenated on 100% oxygen with a PEEP of 6 cmH20.

99

This improvement became more marked when IRV was continued for longer than 3 h, but the change was not in proportion to the duration of IRV. Although an increase in the 1:E ratio from 1:1.9 to 2: 1 produced a significant reduction in Q,/Q,, further reductions in the I:E ratio from 2:l to 2.6:1 or 4:l did not lead to a further significant reduction in Q,/Q,. These beneficial changes in arterial oxygenation corresponded to the changes noted in MAWP, while there were no significant changes in Pace,, systemic or pulmonary hemodynamics, or oxygen delivery. It was previously reported that the effect of IRV on arterial oxygenation was not achieved immediately, but that the improvement became significant with continued IRV [4]. In our study, following the initiation of IRV it generally took more than 3 h to significantly increase the Pao,, and an improvement after around 3 h was seen in only a few patients. Few patients did it take around 3 h to do so. This may explain why some investigators have reported no improvement in Pao, with IRV, since the results were assessed after 15 min [7,8], 30 min [9], and 45-60 min [6] in previous studies. Our results suggested that one explanation for the improvement caused by IRV is alveolar recruitment, with the increase in functional residual capacity (FRC), contributing to improvement of the pulmonary ventilation/perfusion imbalance and the change in MAWP accounting for the improvement in oxygenation. Cole et al. [12] compared the pulmonary and hemodynamic parameters during IRV and conventional ventilation with PEEP that was titrated to achieve a similar external thoracic size. The external thoracic size was increased an average of 1200 ml by IRV and similar changes in external thoracic size could be produced by an average PEEP of 12.8 cmH*O. They suggested that IRV reduced the venous admixture in proportion to the changes in the FRC and is therefore analogous to the use of PEEP. The role of MAWP in improving oxygenation has been debated [5,8,13,14], but our data support the studies by Boros [ 131 and Stewart et al. [ 141demonstrating that an increase in oxygenation appears to be directly related to an increase in MAWP. Prolonging inspiration produces an increase in MAWP with constant and/or low PIP levels. In the present study, the MAWP was significantly higher during IRV than during the control period, which helps to confirm the suggestion that arterial oxygenation in ARF is a function of MAWP. High levels of PEEP are reported to be associated with an increased incidence of pulmonary barotrauma (PBT) and a safe upper limit for PEEP has not yet been established. Since we found that PIP did not increase with IRV, it might be expected to cause less PBT. However, 2 out of 14 patients developed a pneumothorax during IRV in the present series and required the insertion of a chest tube. No PBT is reported to occur at a PIP of less than 50 cmHzO[ 151, and the PIP in our patients with pneumothorax was less than 40 cmH*O, so a high PEEP level did not seem to be responsible. It is possible that the ventilator caused a transient increase in airway pressure with subsequent alveolar rupture in those patients. The two patients had chronic myeloblastic leukemia and systemic lupus erythematosus, respectively, which were complicated by interstitial pneumonia that induced ARF. It may also be possible, therefore, that interstitial pneumonia is more likely to be associated with pneumothorax whatever the primary disease may be. An increase in the I:E ratio which produces an increase in FRC may reduce CO

100

and impair oxygen delivery to the tissues, despite an increase in the Pao, [16]. In this study, we investigated the optimum prolongation of the 1:E ratio for producing an improvement in arterial oxygenation without reducing the CO to such an extent that there is a reduction in the overall tissue b02. Our results suggested that an optimal 1:E ratio would be 2: 1, which produces better gas exchange without significantly affecting CO. No evidence of tissue hypoxemia was observed, with the Vo,, bo2, and PKrz (which remained above 3 1 mmHg [ 171) being stable in all patients during IRV at various 1:E ratios. These results indicate that total body tissue oxygenation was probably not impaired by IRV. The stiffer a patient’s lungs become, the more pressure is required to inflate them. Therefore, we should carefully manage respirator adjustments by making stepwise prolongations of the 1:E ratio and closely observing the hemodynamics (including measurement of CO). Such an approach should allow the successful use of IRV. Our surviving patients showed a significantly lower APACHE II score than that of those who died, indicating that the individual patient’s diseases influenced the outcome, as would be expected. Although we evaluated at early point in time, such as at the time of ICU admission, APACHE II would make the severity classification more independent from treatment [ 111. In conclusion, our study showed that IRV superimposed on PEEP had a beneficial effect on arterial oxygenation when PEEP alone had failed, and that prolongation of the inspiratory time beyond an 1:E ratio of 2:l was not necessary. Also, a significant effect of IRV on arterial oxygenation took longer than 3 h to occur. Although these results only apply directly to our small series, it seems possible that IRV may be useful in patients with ARF. REFERENCES E.O.R. change

Reynolds, in hyaline

Effects of alterations in mechanical ventilation settings on pulmonary membrane disease, Arch. Dis. Child., 46 (1971) 152-159.

gas ex-

E.O.R. Reynolds and A. Taghizaden, Improved prognosis of infants mechanically ventilated for hyaline membrane disease, Arch. Dis. Child., 498 (1974) 505-515. M. Baum, H. Benzer, N. Mutz, G. Pauser and L. Tonczar, Inversed ratio ventilation in der Beatmung beim ARDS, Anaesthesist, 29 (1980) 592-596. S. Duma, H. Baum, H. Benzer, W. Keller, N. Mutz and G. Pauser, Inversed ratio ventilation (IRV) nach kardiochirugishen Eingriffen, Anaesthesist, 3 I (I 982) 549-556. M.J. Gurevitch, J. Van Dyke, E.S. Young and K. Jackson, Improved oxygenation and low peak airway pressure in severe adult respiratory distress syndrome. Treatment with inverse ratio ventilation, Chest, 89 (1986) 211-213. SF. Fuleihan, R.S. Wilson and H. Pontoppidan, Effect of mechanical ventilation with endinspiratory pause on blood-gas exchange, Anesth. Analg., 55 (1976) 122-130. DC. Tyler and F.W. Cheney, Comparison of positive end expiratory pressure and inspiratory positive pressure plateau in ventilation of rabbits with experimental pulmonary edema, Anesth. Analg., 58 (1979) 288-292. L.S. Berman, J.B. Downs, A. Van Eeden and D. Delhagen, 1nspiratory:expiratory ratio. Is mean airway pressure the difference? Crit. Care Med., 9 (1981) 775-777. R.D. Perez-Chada, R.G. Gardaz, R.G. Madgwick and M.K. Sykes, Cardiorespiratory effects of an inspiratory hold and continuous positive pressure ventilation in goats, Intensive Care Med., 9 (1983)

263-269.

101 10 W.A. Knaus, E.A. Draper, D.P. Wagner and J.E. Zimmerman, APACHE II: A severity of disease classification system, Crit. Care Med., 13 (1985) 818-829. 11 J.H. Comroe, R.E. Forester, A.B. Dubois, W.A. Brisco and E. Carlsen, The lung: Clinical Physiology and Pulmonary Function Tests, Year Book Medical Publishers, Chicago, 1962. 12 A.G.H. Cole, SF. Weller and M.K. Sykes, Inverse ratio ventilation compared with PEEP in adult respiratory failure, Intensive Care Med., 10 (1984) 227-232. 13 S.J. Boros, Variation in inspiratory: expiratory ratio and airway pressure wave form during mechanical ventilation: The significance of mean airway pressure, J. Pediatr., 94 (1979) 114-I 17. 14 A.R. Stewart, N.N. Fine and K.L. Peters, Effects of alterations of inspiratory and expiratory pressure and inspiratory/expiratory ratios on mean airway pressure, blood gas, and intracranial pressure, Pediatrics, 67 ( 1981) 474--l8 1. 15 G.W. Petersen and H. Baier, Incidence of pulmonary barotrauma in medical ICU, Crit. Care Med., 11 (1983) 67-69. 16 J.S. Lutch and J.F. Murray, Continuous positive-pressure ventilation and tissue oxygenation: effects on systemic oxygen transport and tissue oxygenation, Ann. Intern. Med., 76 (1972) 193-202. 17 K. Shibutani, T. Komatsu, K. Kubal, V. Sanchala, V. Kumar and D.B. Bizzarri, Critical level of oxygen delivery in anesthetized man, Crit. Care Med., 11 (1983) 640-643.